This project investigates the role of automatic cognitive processes, as compared with conscious, effortful processes, in the memory of mentally retarded persons. A current theoretical position holds that the relationship between short-term memory and intelligence level is due to effortful memory processes. Deficits in these processes in retarded persons account, in part, for their low intelligence. More recent evidence suggests that automatic processes may be involved as well. Automatic processes are of two types, acquired and organic (structural). A series of studies will attempt to define the role of both types of automatic processes and to determine the extent to which nonretarded, cultural-familial retarded, and brain-injured retarded persons differ in these processes. Further, the effects of extended practice on learned automatic processes will be assessed. It is expected that brain injured retarded persons will show deficits in structural processes. Both types of retarded persons will show deficits in learned processes as contrasted with nonretarded persons. The ultimate purpose of this research is to explain short-term memory failures in mentally retarded persons.